Evidence of glutamatergic cells was reported in non-guided protocols of cerebral organoids (COs) (2.5 months) when frequent calcium spikes were observed after the addition of exogenous glutamate (Lancaster et al., 2013). In 3-month-old cerebral organoids, excitatory cortical neurons comprised almost 50% of the cells, shifting to ~23% after 4 months (Fair et al., 2020). Glutamatergic synapses were identified in 6-month-old human cortical spheroids by the co-localization of either vGluT-1 with synapsin (SYN-1) or NMDA receptor subunit NR2B with postsynaptic protein PSD-95 (PSD-95) visualized by array tomography. These cortical glutamatergic neurons displayed a layer-specific organization (Paşca et al., 2015). The presence of glutamatergic synapses was confirmed by the co-localization of pre and postsynaptic markers VGluT1 and PSD95 in dorsal forebrain organoids with 3- and 6-months (Velasco et al., 2019).

In guided protocols, glutamatergic neurons (positive for GLUD1, vGLUT1, and vGLUT2) were highly enriched in 1-month brain organoids resembling cortical domain (Xiang et al., 2017). Similarly, glutamatergic neurons (vGluT1) were detected in 1-month cortical organoids, reaching a substantial portion of the neuronal population between 3 and 6 months in culture, with some of them co-expressing subunits of GABAergic receptors (Trujillo et al., 2019). Similar to GABA transcripts, glutamatergic markers were highly expressed in 1–1.3 month Bioengineered Neuronal Organoids (BENO's) as well as cortical neurons positive for vesicular glutamate transporter (vGLUT) and receptor (GLUR1) (Zafeiriou et al., 2020).

Single-cell profiling analysis revealed that glutamatergic genes were also highly expressed after 2 months of culturing of human subpallium spheroids (Birey et al., 2017), with a predominance of 55% of the total cells in 3.5-month human cortical spheroids encompassing glutamatergic neurons co-expressing cortical layer markers (Yoon et al., 2019). A small group of glutamatergic neurons, identified by genes encoding glutamate transporters (SLC17A7 and SLC17A6), was identified in 2.5-month striatal spheroids, but their expression was more correlated with the neocortex and amygdala (Miura et al., 2020). Overall, glutamatergic neurons have been abundantly observed in cortical organoids.

Calretinin-positive cells have been observed among neuronal types found in self-organized COs (obtained with non-guided protocols which did not display the identity of a specific brain region). They have a typical morphology of tangential migration neurons that is suggestive of interneurons. Also, GABA positive (vGAT) cells were observed in regions resembling medial ganglionic eminence (MGE) (Lancaster et al., 2013; Renner et al., 2017). Immunohistological analysis in 3-month-old forebrain organoids showed GABAergic neurons expressing VGAT, parvalbumin, nNOS, or somatostatin (Qian et al., 2016). By culturing slices from COs in an air-liquid interface, named (ALI-CO), Giandomenico and collaborators showed an improvement in cell viability and maturation when compared to previous non-guided protocols. This protocol also led to higher levels of GABA and expression of inhibitory synapse formation genes in 2.5 months-old ALI-CO (Giandomenico et al., 2019). In a recent study, single-cell RNA sequencing comparison of COs at two time points found 1.8% of cells to be interneurons after 3 months in culture—a number that reached 30.4% at 4.3 months. Besides, the expression of GABA (parvalbumin or somatostatin) for mature GABAergic neurons increased in cerebral organoids and acquired a mature phenotype (larger cell bodies) between 3 and 5 months in culture (Fair et al., 2020). A similar analysis had been performed in non-guided COs that identified the expression of both cortical GABAergic interneuron genes (including markers DLX5, DLX2, SCGN, and GAD1) and genes of GABAergic synapses by 6-months in culture (Quadrato et al., 2017). Even though GABAergic interneurons originate from ventral forebrain progenitors, some GABA genes and markers may be found in cerebral organoids, whose cell diversity encompasses mainly regions of the rostral forebrain.

GABAergic neurons were observed in brain organoids exposed to dorsal patterning clues around 2.5 months of culturing (Xiang et al., 2017). In a more detailed analysis of dorsal spheroids, including bulk RNA sequencing, a small proportion of GABAergic neurons was found in human cortical spheroids at 3.5 months, which may be assigned to a minor ventralization that has eventually occurred (Yoon et al., 2019). Trujillo et al. used guided protocols, specific to generate region-specific organoids, with cortical specification (Paşca et al., 2015; Yoon et al., 2019). They showed that human cortical organoids should be cultured for 3–6 months before GABAergic neurons could be observed, and for 6 months to detect GABA in the culture medium. Even after long periods of maturation (up to 10 months), GABAergic neurons comprised only 15% of the total population of neurons (Trujillo et al., 2019). Thus, the presence of GABAergic markers occurs to a lesser extent in organoids with a dorsal domain and requires more time of culture, as a predictable pattern resembling human brain development.

Studies comparing different species showed that 40% of supragranular layers, among all cortical neurons, were identified as interneurons in primates vs. ~15–20% in rodents (Defelipe, 2011). Likewise, 5% of neurons are interneurons in the mouse thalamus, while in humans, they represent 30% (Arcelli et al., 1997). Since interneurons mainly migrate from the ventral telencephalon, some groups aiming to model interneuron migration have developed protocols guided to either the ventral (subpallium) or dorsal (pallium) forebrain. In vivo, the ganglionic eminences (GE) generate interneurons that migrate tangentially into the dorsal forebrain (Hansen et al., 2013; Ma et al., 2013). As a result, an enrichment of GABAergic markers was identified according to the specification of brain regions in a shorter time in culture. In brain organoids of the medial ganglionic eminence, the expression of GABA synthesis enzymes and vesicular GABA transporter (vGAT) genes were highly enriched after 1 month, while GABAergic synapses were confirmed by immunohistochemistry after 2 months in culture (Xiang et al., 2017). Likewise, in human subpallium spheroids, GAD67 (GABA enzyme synthesis), somatostatin, calretinin, calbindin, and GABA per se were already observed within 2 months. Single-cell profiling identified a group of GABAergic cells (DLX1, GAD1, SLC32A1, SCG2, SST) at 3.5 months, while mature cells positive for parvalbumin were described only after 6 months of culturing (Birey et al., 2017). The fusion of ventral and dorsal cerebral organoids presented GAD1 or vGAT positive cells in both regions at 2.6 months and were dorsally stained for GAD1 cells migrating from the ventral region (Bagley et al., 2017).

To model cortico-striatal circuits of the forebrain, spheroids resembling the lateral ganglionic eminence were differentiated to give rise to human striatal spheroids. Single-cell RNA-sequencing analyses identified more than 50% GABAergic genes (STMN2+, SYT1+) and GABA-synthesizing enzymes (GAD1 and GAD2) in these organoids at 2.5 months. Immunostaining for calbindin, GAD67, and other markers indicated the presence of medium spiny neurons in the striatal spheroid (Miura et al., 2020). In human midbrain-like organoids, GABA-positive mature neurons were detected after 1 month in culture (Jo et al., 2016), similar to the observed in BENOs, which showed transcripts of GABA as well as cells positive for GABA receptors around 1 month in culture (Zafeiriou et al., 2020).

Primary microcephaly is characterized by a reduction of brain size caused by insufficient production of neural progenitors and neurons during fetal development (Jayaraman et al., 2018). Primary microcephaly was elegantly modeled in the 2013 whole-brain organoid protocol from Lancaster (Lancaster et al., 2013). In that work, iPSCs from patients with microcephaly formed smaller embryoid bodies and showed reduced neuroepithelial tissue at an early stage of organoid development (22-days) (Figure 2).

Inherited mutations or external perturbations can lead to microcephaly mainly by the disruption of cell division-related processes (Jean et al., 2020). Brain organoids display the positioning of radial glia-like cells in an orientation that resembles that of a ventricular zone facilitating the study of oriented cell divisions. This way, the pattern of symmetrical/asymmetrical divisions in humans could be studied in vitro (Lancaster et al., 2013). Symmetrical divisions account for the expansion of neural progenitors, whereas asymmetric divisions generate neurons. Increased asymmetric divisions anticipate the generation of neurons and result in reduction of cortical size (Matsuzaki and Shitamukai, 2015).

Mutations in genes involved in cell division were shown to produce organoids with reduced size at an early stage of development. CDK5RAP2 participates in cell division coordinating microtubule dynamics as it has microtubule-organizing centers. iPSCs from patients with CDK5RAP2 mutation were used to produce whole-brain organoids. Loss of function of CDK5RAP2 protein resulted in a decrease in symmetrical cell divisions, and reduction of neurogenic regions and premature neurogenesis was observed as a marked increase in the number of BrdU+/doublecortin (DCX)+ cells in patient organoids (Lancaster et al., 2013).

Mutation in the Aspm gene (abnormal spindle-like microcephaly-associated) is the most common cause of human primary microcephaly. iPSCs derived from a patient with a mutation in the Aspm gene formed smaller embryonic bodies compared to control. Reduction in neuroepithelial regions was observed in cortical organoids together with reduced staining for Pax6, Sox2, and ZO-1 as well as fewer proliferative Ki67+ and pH3+ cells. On day 65 of development (~2 months), Aspm mutant organoids showed disorganized lamination as seen by Tbr2 immunostaining. Calcium signaling could be observed at day 85 (2.8 months) of development and showed no differences in transient frequencies, but mutant organoids were less synchronized than controls, suggesting that those cells were not able to develop mature circuits (Li et al., 2017).

The future characterization of other mutations leading to microcephaly can amplify the knowledge about the developmental processes that guide brain size [reviewed in Gabriel et al. (2020)]. Also, it opens possibilities to investigate how the establishment of a neural progenitor pool has a role in circuit formation. Reduced organoid size in response to pathogens has also been described and will be discussed in the topic: organoids as a model to study infectious disease.

Rett syndrome (RTT) is a rare neurodevelopmental disorder characterized by speech and motor disabilities, along with cognitive impairment, which start to appear after a period of normal development. Mutations in the methyl-CpG-binding protein 2 (MeCP2) gene are the cause of Rett syndrome, and the spectrum of severity can vary depending on the site of mutation. Since the MeCP2 gene is located in the X chromosome, female cells present a mosaic pattern of expression (Chahrour and Zoghbi, 2007).

Rett syndrome was studied using iPSC cells from patients carrying mutations in the MeCP2 gene. Neuronal cultures differentiated from patients' iPSCs allowed the characterization of an in vitro phenotype showing reduced synapses, altered calcium signaling, defective firing activity, and excitatory/inhibitory imbalance (Marchetto et al., 2010). MeCP2 is critical for the functioning of gabaergic neurons (Chao et al., 2010). Using neurons derived from iPSCs from RTT patients, it was possible to show that downregulation of a neuron-specific K+-Cl– cotransporter2 (KCC2) was the cause of GABA functional deficits in Rett neurons, impeding GABA functional switch (Tang et al., 2016). Because KCC2 starts to be expressed postnatally and increases throughout the lifespan, deficits in KCC could explain why the disease starts to manifest early in childhood and progressively increases severity with aging.

Studies addressing 3D whole-brain organoid models of RTT showed deficits in neurogenesis, with an increase of ventricular zones at 1.15 months (5 weeks) of development. Increased proliferation in detriment of neuronal generation was suggested as the cause of delayed development observed in 2D neuronal cultures (Mellios et al., 2018). MEA electrophysiology conducted in MECP2-KO cortical organoids showed decreased population spiking compared with the controls, which could be rescued with treatment with selected compounds from a drug screening (Trujillo et al., 2021). Region-specific organoids with cortical specialization and medial ganglionic eminence (MGE), a critical ventral brain domain producing cortical interneurons and related lineages (Xiang et al., 2017), were generated from hESC with MeCP2 mutation. Calcium signaling and depolarization-dependent c-FOS induction were analyzed in wt and MeCP2 mutant organoids and showed lack of synchronization of calcium surges and deficient c-Fos induction in MGE organoids. Synchronization of calcium surges was not affected in cortical organoids, although c-Fos induction was compromised, suggesting that cortical areas, isolatedly, are less sensitive to MeCP2 mutation. Transcriptional analysis of 2.4 months (74 days) organoids suggested that potential deficits in GABAergic activity seen in these organoids could be reversed by a drug affecting epigenetic pathways (Xiang et al., 2020).

Patient-derived female iPSCs carrying a MeCP2 mutation were used to generate lines with an isogenic control of the MeCP2 mutation. Dorsal organoids were generated and altered intracellular calcium dynamics were observed in neurons from dissociated organoids at 1.5 months (44–47 days) and 2.2 months (64–67 days). Alterations in neuronal activity were measured using patch-clamp at 1.5 and 2.6 months (81 days), revealing a reduction in the firing frequency and abortive-like action potential in mature MeCP2 mutant neurons. The fusion of region-specific organoids from MGE and cortex revealed defective interneuron migration in MeCP2 mutant assembloids at day 41 (1.3 months) (fusion of assembloids occurred at day 13) (Gomes et al., 2020a). In the future, it will be important to expand the analysis of RTT assembloids to address how dysfunction of interneurons caused by MeCP2 mutation can interfere with circuitry formation and cortical functioning.

Autism spectrum disorder (ASD) is a frequent neurodevelopmental disorder characterized by impairments in social interaction, communication, and behavioral flexibility (DMS-5, American Psychiatric Association, 2013). The genetic determinants of ASD have been largely investigated, and due to the clinical heterogeneity of ASD, understanding how key genes play a role in the establishment of ASD is a challenging task. A recent large-scale exome sequencing study identified 102 risk genes related to autism. Most risk genes are expressed in early brain development and have a role in the regulation of gene expression or in neuronal communication (Satterstrom et al., 2020).

CDH8 (chromodomain helicase DNA-binding protein 8) is a chromatin remodeling factor, which was identified as a top candidate gene for ASD from independent exome-sequencing studies and is also involved in schizophrenia and intellectual disabilities (Neale et al., 2012; O'Roak et al., 2012; Talkowski et al., 2012; Bernier et al., 2014; Krumm et al., 2014; McCarthy et al., 2014). The investigation of the transcriptional profile of neural lineages upon CDH8 knockdown or knockout has been used to understand which genes and pathways are affected in ASD. The sh-RNA knockdown of CDH8 in neural progenitor cells and a neuroblastoma cell line affected genes involved in neurodevelopment and altered the expression of genes related to autism (Bernier et al., 2014; Sugathan et al., 2014; Cotney et al., 2015). CDH8 knockout in an iPSC line was introduced using the CRISPR-Cas9 technology, generating a mutant cell line that could be compared with its isogenic control (Wang et al., 2015). Neural progenitor cells and a mixed neuronal culture positive for markers from distinct brain regions were differentiated from mutant cells and isogenic controls, and regulated genes were involved in cell adhesion, neuron differentiation, neuron projection, synaptic transmission, axonal guidance signaling, and WNT/β-catenin and PTEN signaling (Wang et al., 2015).

Brain organoids were generated from CHDH8^−/−, two heterozygous lines, and two isogenic controls (Wang et al., 2017). Transcriptional analysis at 1.6 months (day 50) revealed a significant overlap (~50%) to genes regulated in CHDH8^−/− neural progenitor cells and neurons. DLX6-AS1 and DLX1, which regulate gabaergic neuron development, are among the top regulated genes in CHDH8^−/− brain organoids. DLX6-AS1 was also one of the top genes found in a transcriptome study of brain organoids from idiopathic ASD patients, which were analyzed on days 11 and 31 in vitro (Mariani et al., 2015). Other genes involved in brain development, and previously implicated in ASD, were found to be regulated in CHDH8^−/− and idiopathic ASD, including FZD8, PAX6, SLC1A3, EOMES (TBR2), and MPPED1 (Mariani et al., 2015; Wang et al., 2017), showing that distinct genetic alterations can lead to convergent alterations in gene expression.

The transcriptional analysis of brain organoids from ASD probands revealed alterations in cell cycle genes. The cell cycle in ASD was reduced in length in comparison to controls, resulting in increased neuronal differentiation and overproduction of neurons, and the ratio between glutamatergic and gabaergic neurons was altered (Mariani et al., 2015). Sodium currents in ASD neurons from brain organoids tended to inactivate at more hyperpolarized membrane potentials than in controls. Half maximal amplitude values in the range −72 to −65 mV suggested increased expression of the Nav1.1 isoform, which is preferentially expressed in GABAergic interneurons (Ogiwara et al., 2007) and consistent with a greater proportion of GABAergic neurons in ASD cortical organoids. Importantly, FOXG1 was upregulated in idiopathic ASD, but not CHDH8-/- brain organoids, and ASD phenotype could be reverted by FOXG1 knockdown (Mariani et al., 2015).

The importance of early steps in development to the establishment of ASD phenotypes was demonstrated in brain organoids from patients with ASD. Quantitative assessment of neuronal branching in 6-week organoids showed aberrantly complex neurite structures in ASD neurons when compared to controls. Notably, circumventing the neural stem cell step of development by using a directed differentiation protocol avoided the ASD phenotype (Schafer et al., 2019). This study highlighted the importance of models that recapitulate early stages of development to the comprehension of ASD phenotype.

Moreover, ASD risk genes are enriched in excitatory and inhibitory neuronal lineages, and a combination of gene expression profiles could promote the excitatory-inhibitory imbalance underlying ASD (Satterstrom et al., 2020). It would be interesting, in the future, to address how risk genes can affect brain organoid circuitry at the electrophysiological level and how it correlates to ASD phenotypes.

ZIKV is a mosquito-borne flavivirus that became a global health concern after the connection between infection during pregnancy and the occurrence of microcephaly in children born from infected mothers was established (www.who.int). ZIKV experimental infection in brain organoids induced cell death and reduced proliferation. Also, a decrease in neuronal cell-layer volume and in organoid size resembling microcephaly phenotypes was observed following infection (Cugola et al., 2016; Dang et al., 2016; Garcez et al., 2016; Qian et al., 2016). Moreover, ZIKV infection led to mitochondrial failure, oxidative stress, and DNA damage in human iPSC-derived astrocytes and mitochondrial damage in brain organoids (Ledur et al., 2020).

The molecular basis of ZIKV infection was further explored in proteomic and transcriptomic analysis of ZIKV infected neurospheres, a 3D model of neural culture lacking cortical layer organization, revealing alterations in molecular pathways regulating cell cycle, neuronal differentiation and innate immune response (Garcez et al., 2017). Accordingly, upregulation of the innate immune receptor Toll-Like-Receptor 3 (TLR3) was shown after ZIKV infection of human organoids at early days in vitro (15 days), and inhibition of TLR3 reduced the phenotypic effects of ZIKV infection (Dang et al., 2016).

Infection of human brain organoids has been used for the screening of compounds and drug repurposing to treat infection (Xu et al., 2016, 2019; Watanabe et al., 2017; Pettke et al., 2020). Strikingly, brain organoids were used by our group to establish a connection between ZIKV infection and the regional increase in the cases of microcephaly in the northeast of Brazil. Analysis of the quality of water in reservoirs revealed the presence of saxitoxin in a period concomitant to the appearance of microcephaly cases suggesting that STX and ZIKV could act as a co-insult, aggravating that condition. We showed that treatment with STX enhanced ZIKV phenotype in brain organoids, raising awareness about the impact of water quality in viral infection (Pedrosa et al., 2020).

The strike of COVID-19 pandemic fuelled research in the organoid field to try to understand the neurological symptoms associated with the disease. Although COVID-19 mainly affects the respiratory tract, neurological symptoms are also present in 30–60% of patients, including paresthesia, altered consciousness, and headache (De Felice et al., 2020; Wang et al., 2020). COVID-19 increases the risk of stroke by 7-fold, and 2–6% of patients develop cerebrovascular disease (Fifi and Mocco, 2020). Also, encephalopathy, encephalitis and Guillain-Barré syndrome appear as less frequent neurological manifestations of COVID-19 (Ellul et al., 2020; Mao et al., 2020).

Experimental SARS-CoV-2 infection in brain organoids helped to elucidate types of cells affected by this virus. Cortical organoids at 2 weeks and 1 month were infected with SARS-CoV-2 and fewer SARS-CoV-2 positive cells were identified at 2 weeks than 2 months, revealing a preference for the infection of mature neurons. Also, neurons from brain organoids in an air-liquid interface culture, plated at 2 months and migrated for 2 weeks, were positive for SARS-CoV-2. In all systems analyzed, infection was not shown to be productive. Two-month-old organoids positive for SARS-CoV-2 exhibited an altered Tau localization pattern from axons to soma, Tau hyperphosphorylation, and apparent neuronal death (Ramani et al., 2020). Neuronal infection was shown by other groups (Mesci et al., 2020; Song et al., 2021) and productive infection was observed using high multiplicity of infections (MOI 10). However, the comparison of the conditions used to infect neurosphere cultures using a different SARS-CoV-2 isolate could not confirm productive infection, suggesting that differences between strains can affect neurotropism (Pedrosa et al., 2021). SARS-CoV-2 infection could be prevented by blocking ACE2 with antibodies, and by using cerebrospinal fluid from a COVID-19 patient (Song et al., 2021). Analysis of ACE2 expression in the brain revealed much higher expression in the choroid plexus than in the cortex (Pellegrini et al., 2020), and a much higher SARS-CoV-2 tropism to choroid plexus was identified, but little to no infection of neurons or glia was shown (Jacob et al., 2020; Pellegrini et al., 2020). Infection of choroid plexus organoids was demonstrated (Jacob et al., 2020; Pellegrini et al., 2020) and cell death and transcriptional dysregulation indicative of an inflammatory response and cellular function deficits was seen (Jacob et al., 2020). Leakage in choroid plexus organoids was observed, suggesting a functional loss in the barrier function of this tissue in vitro (Pellegrini et al., 2020), which could make brain tissue vulnerable to the entry of pathogens and inflammatory molecules. These findings were corroborated with post-mortem studies showing SARS-CoV-2 infection in the choroid plexus (Gomes et al., 2020b).

Recently, assembloids constituted by pericytes and cortical organoids were shown to facilitate the infection of astrocytes by SARS-CoV-2. Astrocytic cell death was shown along with increased cell stress and inflammation (Wang et al., 2021). Given that COVID-19 severely damages the brain vascular system (Lee et al., 2021a), vascularized organoids could contribute to understanding the role of the vascular tissue damage in brain infection. Additionally, since there is increasing evidence of neurological and neuropsychiatric complications of COVID-19 (Varatharaj et al., 2020; Ding et al., 2021), it would be interesting to study SARS-CoV-2 infection in models to replicate diseases that may lead individuals most vulnerable to neurological COVID-19 sequelae.

AD is characterized by progressive neurodegeneration and memory loss and is the most prevalent cause of dementia accounting for 60–80% of cases. The presence of senile plaques, constituted by amyloid-β peptides (Aβ) and neurofibrillary tangles (NFTs), formed by phosphorylated tau, are two major neuropathological hallmarks in AD which often precede the onset of symptomatic dementia by decades (Scheltens et al., 2016). In addition, brain stress signaling and glucose metabolism are altered in AD along with a pro-inflammatory profile contributing to AD pathophysiology (Clarke et al., 2015).

Familial AD (FAD) accounts for <5% of AD cases and appears at an early onset (~40 years old) when compared to sporadic AD (SAD), which accounts for the majority of cases and develops later in life (>60 years old) (Paul Johns, 2014). FAD is caused by variants in the presenilin-1 (PSEN1), presenilin-2 (PSEN2), or amyloid precursor protein (APP) genes, whereas SAD is associated with multiple genetic polymorphisms. Known causal variants include the ε4 haplotype in APOE, the strongest genetic risk factor for late-onset AD (Lanoiselée et al., 2017; Schwartzentruber et al., 2021). Also, because APP is located in C21, a connection between gene dosage and AD is seen in 21 trisomy, and Down syndrome adults have increased risk to develop AD and show neuropathological changes of AD at age 40 (Wiseman et al., 2015; Lott and Head, 2019).

AD pathology could be replicated in iPSC-derived neurons from AD patients showing an increase in Aβ, phosphorylated tau, and cellular stress markers (Yagi et al., 2011; Kondo et al., 2013; Muratore et al., 2014). The use of brain organoids allowed for the detection of extracellular protein aggregation in a complex tissue. Brain organoids obtained from patients with FAD with APP duplication and PSEN1 mutation and from individuals with Down syndrome could replicate Aβ accumulation and tau pathology at 1 month, 2 months, and after 3 months in culture (Figure 2) (Raja et al., 2016; Gonzalez et al., 2018; Alić et al., 2020). Investigation of such models showed that APOE4 allele aggravated tau pathology in brain organoids from patients at 3 months in vitro. In addition, the authors showed that the isogenic conversion of APOE4 to APOE3 allele attenuated the AD-related phenotypes (Zhao et al., 2020).

Neuronal hyperactivity is observed in patients with FAD and SAD (AD), leading to non-convulsive epileptic discharges (Vossel et al., 2013; Palop and Mucke, 2016; Lam et al., 2017). Increased calcium transients could be observed in brain organoids from FAD patients (Park et al., 2018) and asynchronous calcium and enhanced neuronal hyperactivity were observed at 40+3 weeks (Yin and VanDongen, 2021). AD neuronal activity in 2-month-old cerebral organoids from FAD patients plated in MEA displayed an increase in action potential firing rate compared to WT organoids and also showed increased VGLUT1 and decreased VGAT staining (Ghatak et al., 2019). Following these observations, a report from the same group showed hypersynchronous network activity in AD organoids, which could be reverted by the dual-allosteric NMDAR antagonist NitroSynapsin, but not by memantine, a drug approved by the FDA for AD treatment (Ghatak et al., 2020).

The role of microglia in AD pathology was studied in co-culture experiments in which microglia-like cells were co-cultured with brain organoids with APP mutation. APOE3- or APOE4 microglia-like cells were added to 2-month-old AD organoids that displayed Aβ aggregates. After 1 month of co-culture, APOE4 allele exhibited more extracellular Aβ aggregates than those co-cultured with APOE3 microglia-like cells, showing that APOE4 negatively impacts microglial function with a poor ability to clear extracellular Aβ from AD brains, and possibly influencing the brain inflammatory profile (Lin et al., 2018).

So far, most of the contributions from brain organoids to the understanding of AD have focused on FAD. Although FAD replicated multiple aspects of SAD with the possible advantage of being detected at earlier times in brain cultivation, it would be important to understand how models for SAD manifest Alzheimer pathology in vitro. iPSC and CRISPR/CAs-9 technology would allow for the analysis of candidate risk genes and their combinations, possibly setting up a model for SAD (Schwartzentruber et al., 2021). Additionally, metabolic imbalance and the role of immune response in AD brain organoids, both of which are current therapeutic targets for AD (Clarke et al., 2015) have been poorly explored. These aspects should be taken into account when using brain organoids for drug discovery.

PD is characterized by the progressive development of movement impairments, and other non-motor symptoms caused by loss of dopaminergic neurons from substantia nigra (SN) in the midbrain. A pathological hallmark of PD is α-synuclein deposition in Lewy bodies. Accumulation of α-synuclein generates synaptic dysfunction which leads to neurodegeneration (Schulz-Schaeffer, 2010; Stefanis, 2012). A mutation in the SNCA gene coding for α-synuclein was the first to be related to PD. Today, multiple risk genes are associated with PD, and missense mutations in the leucine-rich repeat kinase 2 (LRRK2) gene locus are the most common known causes of late-onset familial and sporadic PD (Di Fonzo et al., 2005; Nalls et al., 2019).

Advances in the development of midbrain organoids allowed to test PD in cultured human tissue (Jo et al., 2016; Smits et al., 2019). Sporadic Parkinson's Disease was modeled in midbrain organoids generated from patients carrying a mutation in the LRRK2 gene and compared to isogenic controls. Control midbrain organoids showed a progressive increase in dopaminergic neurons over time (from day 10 to 70), whereas, in patient-derived midbrain organoids, the development of dopaminergic neurons was impaired. Patient-derived midbrain organoids also showed reduced complexity of midbrain dopaminergic neurons (Smits et al., 2019). Reduced astrocytic activity was shown in 35-day old (1.15 month) midbrain PD organoids, which may be associated with neurodegeneration (Kano et al., 2020). Transcriptomic analysis of brain organoids (at 1.5 and 2 months in culture) derived from iPSC in which LRRK2 mutation was introduced using CRISPR/Cas9 revealed the upregulation of a thioredoxin-interacting protein (TXNIP), which was previously reported to be associated with lysosomal dysfunction in α-synuclein-overexpressed cultures. The knockdown of TXNIP suppressed the aggregation of phosphorylated α-synuclein in lysosomes (Kim et al., 2019). This was the first study to provide a direct functional connection between α-synuclein and LRRK2 through TXNIP endorsing the power of brain organoid technology to investigate molecular mechanisms of PD.

A recent publication described a model of early-onset PD in which a mutation in DNAJ6, described in familial juvenile parkinsonism, was introduced in hESC using CRISPR-Cas9 (Wulansari et al., 2021). DNAJ6 mutant midbrain organoids showed defective wnt signaling at day 15, showing that this mutation has an impact on early neurodevelopment. Less dopaminergic neurons were found, and dopaminergic neuron degeneration was observed in midbrain organoids with DNAJ6 mutation (Wulansari et al., 2021), similarly to previous PD models (Smits et al., 2019). Increased reactive oxygen species (ROS), lower levels of dopamine release, increased α-syn oligomers, mitochondrial and autolysosomal dysfunctions were also observed in mutant midbrain organoids (~2 months) (Figure 2) (Wulansari et al., 2021). Mutant midbrain organoids showed an increase in firing frequencies at 2.6 months (DIV80) (Wulansari et al., 2021). The authors correlate this finding with an increase of intrinsic pacemaker frequency, a neuronal characteristic in progressive PD, which is caused by oxidative impairment of A-type Kv4.3 potassium channels (Subramaniam et al., 2014).

The development of non-invasive methods to address organoid-models for neurodegeneration are paramount to measure maturity and pathological features over time. An example of such technology is a pathology imprinted electrode that can detect alpha-synuclein at femtogram levels, which was used to measure alpha-synuclein in the culture medium of human brain organoids generated from normal and idiopathic PD patients (Lee et al., 2021b).